欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
基于狼群群体智能，模拟狼群捕食行为及其猎物分配方式，抽象出游走、召唤、围攻3种智能行为以及“胜者为王”的头狼产生规则和“强者生存”的狼群更新机制，提出一种新的群体智能算法——狼群算法（wolfpack algorithm，WPA），并基于马尔科夫链理论证明了算法的收敛性。将算法应用于15个典型复杂函数优化问题，并同经典的粒子群算法、鱼群算法和遗传算法进行比较。仿真结果表明，该算法具有较好的全局收敛性和计算鲁棒性，尤其适合高雏、多峰的复杂函数求解。
英文摘要：
Based on swarm intelligence of a pach of wolves,inspiring by their hunting behaviors and distrib utive mode for prey,this paper abstracts three intelligent behaviors, scouting, summoning and beleaguering,and deduces a productive rule for leading wolf that the winner can dominate its all and a renewable mechaism, named survival of the stronger, for a pack of wolves. From the above, a new heuristic swarm intelligent method,wolf pack algorithm （WPA） is proposed. Moreover, this algorithm＇s global prohability convergence is proved based on the theory of Markov chain. Then the proposed algorithm is applied to 15 typical complex function optimiza tion problems, and compared with three classical intelligent algorithms,particle swarm optimization algorithm, artificial fish swarm algorithm and genetic algorithm. Simulation results show that WPA has better global con- vergence and computational robustness, and is especially suitable for solving high-dimension and multimodal function optimization problems.
同期刊论文项目
同项目期刊论文
A New Hybrid Intelligent Algorithm for Fuzzy Multiobjective Programming Problem Based on Credibility
期刊信息
位置：